On Thu, 14 Nov 2024 at 14:47, Konrad Dybcio
<konrad.dybcio@xxxxxxxxxxxxxxxx> wrote:
>
> On 14.11.2024 8:47 AM, Varadarajan Narayanan wrote:
> > Enable QMP USB3 phy support for IPQ5424 SoC.
> >
> > Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@xxxxxxxxxx>
> > Signed-off-by: Varadarajan Narayanan <quic_varada@xxxxxxxxxxx>
> > ---
> > v2: Add 'Reviewed-by: Dmitry Baryshkov'
> > ---
> > drivers/phy/qualcomm/phy-qcom-qmp-usb.c | 3 +++
> > 1 file changed, 3 insertions(+)
> >
> > diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-usb.c b/drivers/phy/qualcomm/phy-qcom-qmp-usb.c
> > index acd6075bf6d9..f43823539a3b 100644
> > --- a/drivers/phy/qualcomm/phy-qcom-qmp-usb.c
> > +++ b/drivers/phy/qualcomm/phy-qcom-qmp-usb.c
> > @@ -2298,6 +2298,9 @@ static int qmp_usb_probe(struct platform_device *pdev)
> >
> > static const struct of_device_id qmp_usb_of_match_table[] = {
> > {
> > + .compatible = "qcom,ipq5424-qmp-usb3-phy",
> > + .data = &ipq9574_usb3phy_cfg,
> > + }, {
>
> If the software interface is the same, can this just use ipq9574 as a
> fallback compatible?
Generally I'd agree here, but as PHY tables include not just setup
values, but also platform and chip-specific tunes, I think it's better
to have multiple entries rather than having to cope with the possible
issues. The only "fallback" compatibles that we have are qcs615-ufs ->
sm6115-ufs and qcs8300-ufs -> sa8775p-ufs. Thus I think it's better to
stay within the single-compat model.
--
With best wishes
Dmitry
